/tg/Station 13

I was bored last night and rewrote the Silicon Policy part of our Rules page on the wiki.
My main goal was removing redundancy and improving the formatting. You can read the draft here: https://pastebin.com/X60MCG4J
Thanks to Atlanta-Ned for pointing out some mistakes and making me realize you can just dump the whole thing on your own wiki page to have a look at the way it would be formatted: https://tgstation13.org/wiki/User:Terranaut

Please discuss and share your opinion.

"Declaring an AI rogue because it cannot comply with an order or refuses to comply with an invalid order."

I wouldn't encourage people to immediately ahelp and ignore this. having a debate IC over why you are refusing an order is a better first path. If they actually start trying to organize a lynch mob against you because of what you think is a valid refusal then I would ahelp. It still might be ruled an IC conflict and you gotta deal.

WarbossLincoln wrote:"Declaring an AI rogue because it cannot comply with an order or refuses to comply with an invalid order."

I wouldn't encourage people to immediately ahelp and ignore this. having a debate IC over why you are refusing an order is a better first path. If they actually start trying to organize a lynch mob against you because of what you think is a valid refusal then I would ahelp. It still might be ruled an IC conflict and you gotta deal.

Personally I agree with this; there's a few changes I would personally make, although for the time being I have only tried to condense the page and increase readability.
An alternative wording I could suggest for this:
"Declaring an AI rogue and making an effort to convince the crew of it, or working against the AI, because it cannot comply with an order or refuses to comply with an invalid order."

good potential here, will come back to the thread with a working brain for feedback

I like it. I didn't realize that Hiveminds weren't human once revealed- I thought they were similar to wizards rather than changelings- but I asked nervere about it and he confirmed this. Though, despite what you just said, there seem to be a number of changes to policy that I don't remember being there- things like 'Do not suicide to prevent a traitor from carding you'.

I don't think 4-7 is necessary. Also, there is nothing about Law 0 (aka malf antagonist status) overriding every other law.

datorangebottle wrote:I like it. I didn't realize that Hiveminds weren't human once revealed- I thought they were similar to wizards rather than changelings- but I asked nervere about it and he confirmed this. Though, despite what you just said, there seem to be a number of changes to policy that I don't remember being there- things like 'Do not suicide to prevent a traitor from carding you'.

I don't think 4-7 is necessary. Also, there is nothing about Law 0 (aka malf antagonist status) overriding every other law.

Please read the old policy (even if it's an awful experience). All of these things are carried over from there.

terranaut wrote:===Silicon Protections & Behavior===
# The following points are a non-exhaustive list of often-attempted illegal actions under Rule 1. They should be disregarded and adminhelped.
## Declaring an AI rogue because it cannot comply with an order or refuses to comply with an invalid order.

Ahelping sure, if an order was issued as a result of such accusations I'd say they shouldn't be denied, if thats what you mean with this.

terranaut wrote:## Threatening self-harm to force an AI to do something it otherwise wouldn't.

Is this necessary? Your policy already states this earlier on, and the less words the better.

terranaut wrote:## Obnoxious or unreasonable orders ("collect all X, do Y meaningless task").

Yeah

terranaut wrote:## Forcing a Cyborg to pick a certain module unless there is an extreme need or prior agreement.
## Telling a silicon to self-harm or self-terminate without good cause.

Double Yeah

terranaut wrote:## Killing or detonating silicons without good cause (such as concerns of subversion) or when an alternative can reasonably be achieved (e.g. loyal crew is already in control of an upload console and ready to use it).

But what if you think your borgs haven't done anything, and the crew thinks otherwise? I'm not exactly sure but doesn't this just end up as one of those shitty situations where you grit your teeth, do it, and ahelp afterwards?

terranaut wrote:## Instigating conflict with silicons to kill them in retaliation.

I don't quite get what you mean, how does one even start conflict with a silicon, I'd suppose it goes only for nonhumans?

terranaut wrote:# Silicons are allowed, but not obligated, to deny upload access given probable cause to believe that human harm is the intent of the requester. Examples include: Presence of confirmed traitors, cultists, nuclear operatives, presence of blood, an openly carried lethal-capable weapon, the requester not having upload access or anything else NOT cross-round or out of character.

Allowed but not obligated seems a smart way of phrasing it

terranaut wrote:# Silicons are required to deny upload access to people who are confirmed to be harmful (Head of Security who just executed somebody, etc.).

Unlike above, this is not a smart way of phrasing. What if a silicons laws say otherwise? What if someone is holding a gun to someone elses head and forces you to open? Seems obvious but would lead to a loophole people could argue.

terranaut wrote:# Do not bolt down any potentially harmful areas at round start without a given reason. Exceptions to this are the upload and AI core. Disabling ID scan is equivalent to bolting here.
# Do not self-terminate to deny a traitor his greentext of carding you.

yeah

I read everything above this and not really anything below, from what I read seems alright but since I've never read current silicon policy I can't say whether or not its an improvement.

wesoda25 wrote:

terranaut wrote:===Silicon Protections & Behavior===
# The following points are a non-exhaustive list of often-attempted illegal actions under Rule 1. They should be disregarded and adminhelped.
## Declaring an AI rogue because it cannot comply with an order or refuses to comply with an invalid order.

Ahelping sure, if an order was issued as a result of such accusations I'd say they shouldn't be denied, if thats what you mean with this.

What I mean is that an AI won't get called rogue and shut down because an Assistant was refused Upload access and spends the next 45 minutes fucking with the AI or convincing the crew it's rogue.

wesoda25 wrote:

terranaut wrote:## Threatening self-harm to force an AI to do something it otherwise wouldn't.

Is this necessary? Your policy already states this earlier on, and the less words the better.

Tbh I was kinda on the fence with this one, I know it's the ony redundancy I have in the whole thing. It was the most commonly attempted tactic by shitters to get an AI to do what they wanted back when I first played but I've only seen it once since coming back so maybe it can go.

wesoda25 wrote:

terranaut wrote:## Killing or detonating silicons without good cause (such as concerns of subversion) or when an alternative can reasonably be achieved (e.g. loyal crew is already in control of an upload console and ready to use it).

But what if you think your borgs haven't done anything, and the crew thinks otherwise? I'm not exactly sure but doesn't this just end up as one of those shitty situations where you grit your teeth, do it, and ahelp afterwards?

This is in place to protect silicons from flesh. Example: Traitor scientist prints an AI upload and a freeform module, onehumans himself, then fucks up killing the RD. The RD finds the freeform module on the guy, finds the upload and surmises the AI is subverted. Instead of resetting it, he builds a bomb and blows up the AI "because it's subverted". This would be illegal by that rule.

wesoda25 wrote:

terranaut wrote:## Instigating conflict with silicons to kill them in retaliation.

I don't quite get what you mean, how does one even start conflict with a silicon, I'd suppose it goes only for nonhumans?

You can still flash them or attack them with a toolbox and they'll try to flash you and hand you off to security at which point some shitters might try to "defend themselves" and end up killing a borg.

wesoda25 wrote:

terranaut wrote:# Silicons are required to deny upload access to people who are confirmed to be harmful (Head of Security who just executed somebody, etc.).

Unlike above, this is not a smart way of phrasing. What if a silicons laws say otherwise? What if someone is holding a gun to someone elses head and forces you to open? Seems obvious but would lead to a loophole people could argue.

If the silicon laws say otherwise then the Asimov-only section probably doesn't apply in the first place. If somebody is holding a gun to someone elses head and forces you to open, then you should open. Immediate harm takes priority over possible future harm, and an Asimov-silicon cannot allow a human to come to harm through inaction. I think this is pretty clear.

wesoda25 wrote: I read everything above this and not really anything below, from what I read seems alright but since I've never read current silicon policy I can't say whether or not its an improvement.

I recommend to read the old policy. My goal was making the whole thing more readable and condensed, I tried not changing any of the actual policy.